Why 2 zero light meter positions? Law of reciprocity isn’t being followed by A7rIII

[PS87]

Good day, I have a technical problem. Here’s the situation. I’m in manual mode. F2.8. Changing shutter speed in half stop increments. Camera is pointed at an unchanging scene, consistent light, no movement of the camera. Meter displays:

1/125 is +2.0

1/200 is +1.5

1/250 is +1.0

1/350 is +0.5

1/500 is zero

1/750 is Again zero (not -0.5)

1/1000 is -0.5 (not -1.0)

1/1500 is -1.0 (not -1.5)

The law of reciprocity is lost by the meter.

It doesn’t matter what meter mode or ISO is set (and you know that it shouldn’t). The problem is that Whenever I cross the threshold of zero, there are TWO zero readings. The meter stays at zero a second time even though the shutter speed is changed . It’s the same when changing the aperture reaches zero -there are two zeroes. Why? It doesn’t follow the basic rules. Sony A7r III. No software updates. Thank you in advance.
Puzzled in Chicago

 

[PS87]

Hello Brownie,
Thank you for tackling this problem with me. I'm going to try to make the puzzling behavior of my school's A7rIII somewhat more clear. Essentially, what I'm getting is an incorrect meter read-out when changing the amount of light going into the camera. Please see if your A7rIII does this too.
1). Set M(anual) Mode.
2). Point the camera at a wall that is in light that won't change. I used a black card for the test I made.
3). Set the A7 for half stop increment changes
4). Set any ISO (I used 400 for the test and tested other settings too: 3200, 50 with the same results)
5). Set any aperture (I used f2.8 on the Sony lens).
6). Set any meter mode (I tested all modes with the same results).
7). Set the SS so that the meter reads -2.0
Procedure:
Changing the SS to the next half stop longer will make the meter read -1.5.
The next half stop longer makes it read -1.0.
When you reach ZERO you'll be at 2 stops more light than when you started. Everything's been good up to now.
The next 1/2 stop change should make the meter read +0.5, BUT it stays at ZERO. The actual exposure does change (as it should) but the meter read-out is now incorrect. There is a discrepancy.
The next change is still incorrect at +1.0, and so on.

I tested a second model A7rIII at our school and it behaved in the exact same way.

I've posted the exposures I made during this test to show that the camera is properly changing the amount of light according to the SS setting. It's the read-out that gets "lost."

Please let me know if you're getting the same results.

Thanks!

Paul

 

[PS87]

OK. I'm hoping that someone will try the experiment. It's very strange (misleading) behavior - and it's happening on the two cameras I have access to. I'll get access to two others when school resumes in the new year. I'm an adjunct faculty at the College of DuPage.

 

[spudhead]

Hello Brownie,
Thank you for tackling this problem with me. I'm going to try to make the puzzling behavior of my school's A7rIII somewhat more clear. Essentially, what I'm getting is an incorrect meter read-out when changing the amount of light going into the camera. Please see if your A7rIII does this too.
1). Set M(anual) Mode.
2). Point the camera at a wall that is in light that won't change. I used a black card for the test I made.
3). Set the A7 for half stop increment changes
4). Set any ISO (I used 400 for the test and tested other settings too: 3200, 50 with the same results)
5). Set any aperture (I used f2.8 on the Sony lens).
6). Set any meter mode (I tested all modes with the same results).
7). Set the SS so that the meter reads -2.0
Procedure:
Changing the SS to the next half stop longer will make the meter read -1.5.
The next half stop longer makes it read -1.0.
When you reach ZERO you'll be at 2 stops more light than when you started. Everything's been good up to now.
The next 1/2 stop change should make the meter read +0.5, BUT it stays at ZERO. The actual exposure does change (as it should) but the meter read-out is now incorrect. There is a discrepancy.
The next change is still incorrect at +1.0, and so on.

I tested a second model A7rIII at our school and it behaved in the exact same way.

I've posted the exposures I made during this test to show that the camera is properly changing the amount of light according to the SS setting. It's the read-out that gets "lost."

Please let me know if you're getting the same results.

Thanks!

Paul

Hi you should post this question on the Sony support site and hopefully you will get the answer you are looking for from Sony technical, I have seen this before on other models and just worked around it , cameras are not perfect no matter the brand

 

[spudhead]

Could it be the black card? Are you filling the frame with the card or is there something else in the shot? I can see where the camera may not be noticing a difference. You get to zero as you should and the next shot is also zero. That actually may make sense given the setup, it may not be affecting the meter enough to make a change due to the subject. Have you tried this on a normal scene, like a lit still-life where you can control the light? Something with color/contrast/shading, etc. in the frame.

As you say Tim many variables I see it on the a9 now and again on dark back grounds if you point at scene with variable light colour and shade immediately the camera changes and does what you expect it would, there is more going on it too simple of a test with the bells and whistles on cameras now

 

[PS87]

I've posted the test on youtube. I've joined the Sony forum, per suggestion, and will post things there. Thanks guys!

https://youtu.be/TVputlJ_HGY

 

[PS87]

• Someone at the Sony board did the test and verifies that the A7rIII does behave in the poor manner I describe above. I'm not at all concerned about the actual exposure, I know the meter isn't perfect. What I'm "upset" about is that, because of the error in read-out, I can't use the A7, or, more specifically its camera meter, in a live-time demonstration of 1/2 stop exposure changes in my college classes. The exposure DOES change accordingly, but when the meter read-out fails to follow suit, reporting incorrectly, things fall apart. It is a defect that certainly can be lived with by most photographers, except those that also teach. I guess it's gonna be a Nikon in class. Thanks again.
  

 

[spudhead]

• Someone at the Sony board did the test and verifies that the A7rIII does behave in the poor manner I describe above. I'm not at all concerned about the actual exposure, I know the meter isn't perfect. What I'm "upset" about is that, because of the error in read-out, I can't use the A7, or, more specifically its camera meter, in a live-time demonstration of 1/2 stop exposure changes in my college classes. The exposure DOES change accordingly, but when the meter read-out fails to follow suit, reporting incorrectly, things fall apart. It is a defect that certainly can be lived with by most photographers, except those that also teach. I guess it's gonna be a Nikon in class. Thanks again.

I am not suprised Sony came back with that response, not as it should be but it is what it is and once you are aware of the issue you can work around it if needed, have you done the same test with the Nikon? I know Sony make most of Nikons sensors but I not sure if Nikon make there own meter, I would be interest to hear results on the Nikon in the same test.

 

[PS87]

I think it sounds more like a program issue. I'd still like to know if you get the same results framing something other than a monochrome card.

Watch the video I made.

 

[PS87]

I am not suprised Sony came back with that response, not as it should be but it is what it is and once you are aware of the issue you can work around it if needed, have you done the same test with the Nikon? I know Sony make most of Nikons sensors but I not sure if Nikon make there own meter, I would be interest to hear results on the Nikon in the same test.

Ok, thanks. I'll let you know. It will be awhile since school's on winter break.

 

[PS87]

Thanks, Brownie! I understand and agree with what you said about program modes potentially having "indecision." The problem rears its head in Manual Mode, and it's strictly the "read-out" (the +/-0 displayed) that goes awry. Thanks again for working with me.